Title of article :
Spatial and temporal dynamics of river channel migration and vegetation in central Amazonian white-water floodplains by remote-sensing techniques
Author/Authors :
Peixoto، نويسنده , , Juliana Maerschner Aguiar and Nelson، نويسنده , , Bruce Walker and Wittmann، نويسنده , , Florian، نويسنده ,
Issue Information :
روزنامه با شماره پیاپی سال 2009
Abstract :
We investigated spatial and temporal migration of the Solimões, the Japurá, and the Aranapu River channels in western Brazilian Amazonia with Landsat TM imagery over a 21-year period. Additionally, we classified and monitored how channel migrations affect the distribution of pioneer vegetation and old-growth forest. The cloud-free study area was 153,032 ha — open water plus 3 km inland on each margin. The channel migration rates, expressed as percent dislocation of the open water body of the river year− 1, were lowest in the Japurá River (1.2%), and highest in the Aranapu channel (2.5%), the point bars at river confluence being the most affected landforms subject to geomorphic changes. Annual rates of lateral erosion and accretion of vegetated land along the three rivers were well-balanced. They averaged 0.79 and 0.83% of the cloud-free channel area over the 21 years. The Solimões River was more dynamic than the Japurá River, which can be traced to higher water discharge and sediment load. During the 21 years, the area covered by pioneer vegetation increased by 5.8% of the study area, while late-succession areas decreased by a similar amount (5.5%). According to local biomass estimates of the different vegetation types, these values suggest that C-releases by alluvial erosion would be much higher than C-sequestration caused by the creation of areas suitable for colonization by pioneer vegetation at our study site.
Keywords :
Remote sensing , Wetland mapping , Vلrzea , Alluvial geomorphology , Cecropia forest , Central Amazon Basin , C-sequestration , Vegetation
Journal title :
Remote Sensing of Environment
Journal title :
Remote Sensing of Environment